Questions tagged [change-point]
Methods that attempt to detect when a change occurs in a distribution, process, or function.
473 questions
0
votes
1
answer
70
views
SEM: two time point latent change score model (LCSM) in R using lavaan [closed]
I am fitting an LCSM in R with lavaan, but the model never runs successfully. Can anyone have a look at my code? Thanks a lot.
My latent change score sturcture is like below:
...
2
votes
2
answers
86
views
Best statistical approach to test a threshold effect in a continuous predictor
In medical community, the oxygen–hemoglobin dissociation curve is a classic. It shows the relationship between SpO₂ and PaO₂. Clinically, we often aim to keep SpO₂ above ~90–92% because below that ...
0
votes
0
answers
10
views
Choosing between Yabu–Perron Model 2 vs. Model 3 before applying Kim–Perron unit-root test
I am using the Yabu–Perron (2009) procedure to detect structural breaks in a time series, and I am unsure how to choose between two deterministic specifications. For my series, both models yield a ...
1
vote
0
answers
73
views
Fitting model to data that stops following model at some point
I have data that I know follows $y=-At^2$ up to some point $t_0$, at which point it stops following that model and becomes essentially noise, although with an upward trend. This is an example:
I need ...
- 11
2
votes
1
answer
104
views
Detecting breakpoints in longterm air quality data (bfast, strucchangeRcpp)
I am looking for breakpoints/change points in long-term air pollutant data. Overall, I am trying to figure out how the coronavirus lockdowns have affected the data and whether any changes or ...
- 21
0
votes
1
answer
89
views
Estimate the changepoint where two marginal predictions converge
I have a dataset of participants performing a maximum work task at various loads, under two conditions c("A", "B"). I would like to estimate at what load work becomes the same in ...
- 133
0
votes
1
answer
61
views
Handling recurring convergence issues in multiple change point detection in R
I'm trying to run a series of multiple change point models on a large number of time series variables, but I am running into occasional convergence issues that I would like advice on how to handle. I ...
- 1
1
vote
0
answers
46
views
What change point detection methods are suitable to detect changes in the trend of non-linear univariate time series data?
I have a dataset that consists of outputs from multiple experiments.
Each output is a heteroscedastic univariate non-linear time series.
The domain knowledge is that at the beginning the trend ...
- 11
4
votes
1
answer
71
views
Alternative to M-fluctuation tests
I am interested in testing if parameters in a model change as a function to some other feature. As far as I understand, the fluctuation test can be used Zeileis & Hornik (2007). I wonder what are ...
- 1,647
0
votes
0
answers
63
views
Standard Error of fitted value at breakpoint (segmented regression)
I am currently using the "segmented.lm" function to detect a change point in my data. At this stage I am trying to figure out how to derive the SE of the y value of the corresponding change ...
- 433
2
votes
1
answer
158
views
Understanding the derivation of bayesian online changepoint detection
I am reading bayesian online changepoint detection. Concerned with section 2. I am trying to understand how the algorithm is derived in the section 2. I am finding it difficult to understand the ...
- 2,807
1
vote
1
answer
117
views
Pettitt Test - Multiple changepoints given?
I am using pettitt's test in the trend package and have a few variables where multiple years are given as the changepoint. I know that this is a nonparametric test and am wondering if this is due to ...
0
votes
0
answers
65
views
Trendbreak analysis / Changepoint detection with small dataset?
I have a small data set (2010, 2011, 2012, 2013, 2014) with respective values/frequencies (number of visitors).
The number of visitors increased until 2012 and then decreased abruptly. I want to test ...
- 1
1
vote
0
answers
41
views
changepoint model prediction has a little offset compared to the manual annotation: how to decide the offset to evaluate with f1score?
in changepoint detection models, you want to detect changes of state in timeseries. But the prediction made by the model would be a little different by an offset from the manual annotation. To compute ...
- 11
0
votes
0
answers
110
views
Detection of a single change point / inflection in a survival analysis
I am conducting a survival analysis in patients with and without surgical complications. When examining the Kaplan-Meier curves for these two populations, it becomes evident that surgical ...
